/*
* Sets the head word and the index for an entity, given the parse tree for
* the sentence containing the entity.
*
* This code is no longer used, but I've kept it around (at least for now) as
* reference when we modify preProcessSentences().
*/
private void SetHeadWord(EntityMention entity, Tree tree)
{
IList <Tree> leaves = tree.GetLeaves();
Tree argRoot = tree.JoinNode(leaves[entity.GetExtentTokenStart()], leaves[entity.GetExtentTokenEnd()]);
Tree headWordNode = argRoot.HeadTerminal(headFinder);
int headWordIndex = GetIndexByObjectEquality(leaves, headWordNode);
if (StringUtils.IsPunct(leaves[entity.GetExtentTokenEnd()].Label().Value().Trim()) && (headWordIndex >= entity.GetExtentTokenEnd() || headWordIndex < entity.GetExtentTokenStart()))
{
argRoot = tree.JoinNode(leaves[entity.GetExtentTokenStart()], leaves[entity.GetExtentTokenEnd() - 1]);
headWordNode = argRoot.HeadTerminal(headFinder);
headWordIndex = GetIndexByObjectEquality(leaves, headWordNode);
if (headWordIndex >= entity.GetExtentTokenStart() && headWordIndex <= entity.GetExtentTokenEnd() - 1)
{
entity.SetHeadTokenPosition(headWordIndex);
entity.SetHeadTokenSpan(new Span(headWordIndex, headWordIndex + 1));
}
}
if (headWordIndex >= entity.GetExtentTokenStart() && headWordIndex <= entity.GetExtentTokenEnd())
{
entity.SetHeadTokenPosition(headWordIndex);
entity.SetHeadTokenSpan(new Span(headWordIndex, headWordIndex + 1));
}
else
{
// Re-parse the argument words by themselves
// Get the list of words in the arg by looking at the leaves between
// arg.getExtentTokenStart() and arg.getExtentTokenEnd() inclusive
IList <string> argWords = new List <string>();
for (int i = entity.GetExtentTokenStart(); i <= entity.GetExtentTokenEnd(); i++)
{
argWords.Add(leaves[i].Label().Value());
}
if (StringUtils.IsPunct(argWords[argWords.Count - 1]))
{
argWords.Remove(argWords.Count - 1);
}
Tree argTree = ParseStrings(argWords);
headWordNode = argTree.HeadTerminal(headFinder);
headWordIndex = GetIndexByObjectEquality(argTree.GetLeaves(), headWordNode) + entity.GetExtentTokenStart();
entity.SetHeadTokenPosition(headWordIndex);
entity.SetHeadTokenSpan(new Span(headWordIndex, headWordIndex + 1));
}
}
/// <summary>Find the index of the head of an entity.</summary>
/// <param name="ent">The entity mention</param>
/// <param name="tree">The Tree for the entire sentence in which it occurs.</param>
/// <param name="tokens">The Sentence in which it occurs</param>
/// <param name="setHeadSpan">Whether to set the head span in the entity mention.</param>
/// <returns>The index of the entity head</returns>
public virtual int AssignSyntacticHead(EntityMention ent, Tree tree, IList <CoreLabel> tokens, bool setHeadSpan)
{
if (ent.GetSyntacticHeadTokenPosition() != -1)
{
return(ent.GetSyntacticHeadTokenPosition());
}
logger.Finest("Finding syntactic head for entity: " + ent + " in tree: " + tree.ToString());
logger.Finest("Flat sentence is: " + tokens);
Tree sh = null;
try
{
sh = FindSyntacticHead(ent, tree, tokens);
}
catch (Exception e)
{
logger.Severe("WARNING: failed to parse sentence. Will continue with the right-most head heuristic: " + SentenceToString(tokens));
Sharpen.Runtime.PrintStackTrace(e);
}
int headPos = ent.GetExtentTokenEnd() - 1;
if (sh != null)
{
CoreLabel label = (CoreLabel)sh.Label();
headPos = label.Get(typeof(CoreAnnotations.BeginIndexAnnotation));
}
else
{
logger.Fine("WARNING: failed to find syntactic head for entity: " + ent + " in tree: " + tree);
logger.Fine("Fallback strategy: will set head to last token in mention: " + tokens[headPos]);
}
ent.SetHeadTokenPosition(headPos);
if (setHeadSpan)
{
// set the head span to match exactly the syntactic head
// this is needed for some corpora where the head span is not given
ent.SetHeadTokenSpan(new Span(headPos, headPos + 1));
}
return(headPos);
}